Method and apparatus for determining drivable region information
Embodiments of this application provide a method and an apparatus for determining drivable region information. The method includes obtaining first information, where the first information includes information about an initial drivable region determined based on at least one image, and the at least one image is from at least one camera module. The method also includes obtaining second information, where the second information includes radar detection information. The method further includes determining first drivable region information based on the first information and the second information.
1. A method comprising:
obtaining first information, wherein the first information comprises information about an initial drivable region in a physical region for a vehicle and is based on at least one image of the physical region, and the at least one image is from at least one camera module, the initial drivable region being a region initially identified for the vehicle as free of obstacles to travel, wherein
the information about the initial drivable region comprises information of at least one pixel of an image of the initial drivable region; or
the information about the initial drivable region comprises boundary information of the image of the initial drivable region;
obtaining second information, wherein the second information comprises radar detection information of the physical region;
determining, based on the first information and the second information, whether a region of interest exists in the initial drivable region, wherein the region of interest is a region with an inaccurate detection result based on fusing the first information and the second information;
upon determining that the region of interest exists,
obtaining, based on the first information and the second information, second drivable region information and region information of the region of interest;
sending the region information of the region of interest to the camera module;
obtaining a detection result of the region of interest from the camera module, the detection result of the region of interest indicating whether an obstacle exists in the region of interest; and
determining a first drivable region within which the vehicle is able to travel obstacle-free based on the second drivable region information and the detection result of the region of interest, wherein the first drivable region is within the initial drivable region.
2. The method according to claim 1 , wherein the radar detection information comprises detection information of the initial drivable region.
3. The method according to claim 1 , wherein detecting the region of interest comprises:
classifying image information corresponding to the region of interest.
4. The method according to claim 3 , comprising:
classifying image information corresponding to the region of interest using a semantic segmentation algorithm or an instance segmentation algorithm, or using a classifier.
5. The method according to claim 1 , further comprising:
determining that the region of interest does not exist; and
obtaining second drivable region information based on the first information and the second information, wherein the second drivable region information indicating the first drivable region.
6. The method according to claim 1 , wherein:
the boundary information includes information about f pixels of the image of the initial drivable region corresponding to a boundary of the initial drivable region, and the information about the f pixels is represented as [(u 1 , h 1 ), (u 2 , h 2 ), (u 3 , h 3 ), . . . , (u i , h i ), . . . , (u f , h f )] in a pixel coordinate system, and wherein u i and h i indicate location coordinates of an i th pixel in the f pixels, 1≤i≤f, and f is an integer greater than or equal to 1.
7. The method according to claim 1 , wherein:
the boundary information includes information about r pixels of the image of the initial drivable region corresponding to a boundary of the initial drivable region, and the information about the r pixels is represented as [step, (g 1 , k 1 ), k 2 , k 3 , . . . , k i , . . . , k r ] or [step, (g 1 , k 1 ), k 2 , k 3 , . . . , k i , . . . , k r , step_method] in a pixel coordinate system, wherein g 1 and k 1 indicate location coordinates of a first pixel in the r pixels, k i indicates a vertical coordinate of an i th pixel in the r pixels, 1≤i≤r, a horizontal coordinate of the i th pixel is obtained based on the step, the step indicates an interval step between the horizontal coordinate of the i th pixel and a horizontal coordinate of an (i+1) th pixel, an interval step between the horizontal coordinate of the i th pixel and a horizontal coordinate of an (i−1) th pixel is same as the interval step between the horizontal coordinate of the i th pixel and the horizontal coordinate of the (i+1) th pixel, the step_method indicates a connection relationship between all interval steps, and r is an integer greater than or equal to 1.
8. The method according to claim 1 , wherein:
the boundary information includes information about s pixels of the image of the initial drivable region corresponding to a boundary of the initial drivable region, and the information about the s pixels is represented as [(p 1 , q 1 ), (p 2 , q 2 ), (p 3 , q 3 ), . . . , (p i , q i ), . . . , (p s , q s ), step_method] in a pixel coordinate system, where p i and q i indicate location coordinates of an i th pixel in the s pixels, 1≤i≤s, there is an interval step between a horizontal coordinate of the i th pixel and a horizontal coordinate of an (i+1) th pixel, an interval step between the horizontal coordinate of the i th pixel and a horizontal coordinate of an (i−1) th pixel is different from the interval step between the horizontal coordinate of the i th pixel and the horizontal coordinate of the (i+1) th pixel, the step_method indicates a connection relationship between all interval steps, and s is an integer greater than or equal to 1.
9. The method according to claim 8 , wherein:
the step_method is linear interpolation, quadratic interpolation, cubic spline interpolation, or shape-preserving piecewise cubic interpolation.
10. The method according to claim 1 , wherein
the radar detection information comprises location coordinates information; or
the radar detection information comprises location coordinates information and covariance information of location coordinates.
11. The method according to claim 1 , wherein determining, based on the first information and the second information, whether the region of interest exists in the initial drivable region comprises:
determining a result of fusion using a statistical fusion algorithm based on the first information and the second information.
12. An apparatus comprising: one or more processors, and a non-transitory computer-readable storage medium storing program instructions; wherein, when executed by the one or more processors, the instructions cause the apparatus to:
obtain first information, wherein the first information comprises information about an initial drivable region in a physical region for a vehicle and is based on at least one image of the physical region, and the at least one image is from at least one camera module, the initial drivable region being a region initially identified for the vehicle as free of obstacles to travel, wherein,
the information of the initial drivable region comprises information of at least one pixel of an image of the initial drivable region; or
the information of the initial drivable region comprises boundary information of the image of the initial drivable region;
obtain second information, wherein the second information comprises radar detection information of the physical region;
determine, based on the first information and the second information, whether a region of interest exists in the initial drivable region, wherein the region of interest is a region with an inaccurate detection result based on fusing the first information and the second information;
upon determine that the region of interest exists,
obtain, based on the first information and the second information, second drivable region information and region information of the region of interest;
send the region information of the region of interest to the camera module;
obtain a detection result of the region of interest from the camera module, wherein the detection result of the region of interest indicating whether an obstacle exists in the region of interest; and
determine a first drivable region within which the vehicle is able to travel obstacle-free based on the second drivable region information and the detection result of the region of interest, wherein the first drivable region is within the initial drivable region.
13. The apparatus according to claim 12 , wherein the radar detection information comprises detection information of the initial drivable region.
14. The apparatus according to claim 12 , the instructions further cause the apparatus to:
classify image information corresponding to the region of interest.
15. The apparatus according to claim 12 , wherein
the boundary information includes information about f pixels of the image of the initial drivable region corresponding to a boundary of the initial drivable region, and the information about the f pixels is represented as [(u 1 , h 1 ), (u 2 , h 2 ), (u 3 , h 3 ), . . . , (u i , h i ), . . . , (u f , h f )] in a pixel coordinate system, and wherein u i and h i indicate location coordinates of an i th pixel in the f pixels, 1≤i≤f, and f is an integer greater than or equal to 1.
16. The apparatus according to claim 15 , wherein the instructions further cause the apparatus to:
classify image information corresponding to the region of interest using a semantic segmentation algorithm or an instance segmentation algorithm or using classifier.
17. The apparatus according to claim 12 , wherein
the boundary information includes information about r pixels of the image of the initial drivable region corresponding to a boundary of the initial drivable region, and the information about the r pixels is represented as [step, (g 1 , k 1 ), k 2 , k 3 , . . . , k i , . . . , k r ] or [step, (g 1 , k 1 ), k 2 , k 3 , . . . , k i , . . . , k r , step_method] in a pixel coordinate system, wherein g 1 and k 1 indicate location coordinates of a first pixel in the r pixels, k i indicates a vertical coordinate of an i th pixel in the r pixels, 1≤i≤r, a horizontal coordinate of the i th pixel is obtained based on the step, the step indicates an interval step between the horizontal coordinate of the i th pixel and a horizontal coordinate of an (i+1) th pixel, an interval step between the horizontal coordinate of the i th pixel and a horizontal coordinate of an (i−1) th pixel is same as the interval step between the horizontal coordinate of the i th pixel and the horizontal coordinate of the (i+1) th pixel, the step_method indicates a connection relationship between all interval steps, r is an integer greater than or equal to 1.
18. The apparatus according to claim 12 , wherein
the boundary information includes information about s pixels of the image of the initial drivable region corresponding to a boundary of the initial drivable region, and the information about the s pixels is represented as [(p 1 , q 1 ), (p 2 , q 2 ), (p 3 , q 3 ), . . . , (p i , q i ), . . . , (p s , q s ), step_method] in a pixel coordinate system, where p i and q i indicate location coordinates of an i th pixel in the s pixels, 1≤i≤s, there is an interval step between a horizontal coordinate of the i th pixel and a horizontal coordinate of an (i+1) th pixel, an interval step between the horizontal coordinate of the i th pixel and a horizontal coordinate of an (i−1) th pixel is different from the interval step between the horizontal coordinate of the i th pixel and the horizontal coordinate of the (i+1) th pixel, the step_method indicates a connection relationship between all interval steps, and s is an integer greater than or equal to 1.
19. The apparatus according to claim 18 , wherein
the step_method is linear interpolation, quadratic interpolation, cubic spline interpolation, or shape-preserving piecewise cubic interpolation.
20. The apparatus according to claim 12 , wherein determining that the region of interest exists in the initial drivable region comprises:
determining a result of fusion using a statistical fusion algorithm based on the first information and the second information.